Title :
Overview of traffic flow hybrid ANN forecasting algorithm study
Author :
Zhang, L.D. ; Jia, L. ; Zhu, W.X.
Author_Institution :
Control Sci. & Eng. Sch., Shandong Univ., Jinan, China
Abstract :
ANN is the core of the 2rd and 3rd generation traffic flow forecasting algorithm. In order to master the traffic flow forecasting tendency comprehensively, this paper studied every kind of hybrid artificial neural network forecasting algorithms in detail, including ARIMA ANN model, fuzzy ANN model, GA ANN model, Chaos ANN model, Wavelet Analysis Model, Principal Component Analysis ANN model and Particle Swarm Optimization ANN model. First it abstracted those algorithm flows into common form, then made general network structure, and analyzed every kind network´s virtues and shortcomings, at last, proposed traffic flow prediction developing tendency in future.
Keywords :
fuzzy neural nets; genetic algorithms; particle swarm optimisation; principal component analysis; road traffic; traffic engineering computing; wavelet transforms; ARIMA ANN model; chaos ANN model; fuzzy ANN model; genetic algorithm ANN model; hybrid artificial neural network; particle swarm optimization ANN model; principal component analysis ANN model; traffic flow forecasting algorithm; wavelet analysis model; Artificial neural networks; Biological system modeling; Computational modeling; Gallium; Gallium nitride; Prediction algorithms; Predictive models; ARIMA; FNN; GA; Hybrid ANN; ITS; PSO; chaos; traffic flow forecasting;
Conference_Titel :
Computer Application and System Modeling (ICCASM), 2010 International Conference on
Conference_Location :
Taiyuan
Print_ISBN :
978-1-4244-7235-2
Electronic_ISBN :
978-1-4244-7237-6
DOI :
10.1109/ICCASM.2010.5620414